Remember the days when tape backup was THE way to store data? Wasn't it fun to insert and eject cartridges just to hear the sounds? There were also the labels and barcodes...the entire process was so much of a physical dance. The thing is, tape backup was and probably still is a major part of any IT organization.
A new technological era in backup dawns
Every few decades, there suddenly comes a refreshing approach to a certain process that starts a whole new trend. And indeed, in the early 90’s, Storage Snapshots was introduced to the world, significantly changing data recovery times from days to mere minutes. Now that was a very welcome change! With this development, you could already recover from crashes immediately, without the dreadful wait for tape restores.
Just how terribly long was the wait? Allow me to refresh your memory. Consider a tape library with eight LTO7 tape drives humming along at a maximum throughput of 300MB/sec each, for an aggregated throughput of 8TB an hour. Even at that rate, it may take hours or even days to recover large data sets. In today’s business landscape where the inability to access (or worse, loss of) data can spell disaster, such scenarios will simply not do.
Storage-side snapshots, on the other hand, enable administrators to recover such data sets very quickly, and in the better implementations, with negligible performance impact. Let’s spell this out in more clear-cut terms: while tape backup provided RTO of hours and days, snapshots took the RTO discussion to seconds and minutes.
The need for faster recovery
Even with snapshots however, RPO’s have not changed dramatically. The usual practice of nightly backups have changed to merely a few snapshots a day. With some organizations having literally hundreds or thousands of VMs and applications hosted together, recovering data to a few hours back may not be enough. For them, this could mean serious loss of data and productivity in some cases, and in more unfortunate circumstances, even loss of real business.
Take a manufacturing floor as an example: If a system failure happens in the afternoon, a recovery to that morning’s snapshot will be needed. This will mean that all orders since the morning will have to be reviewed manually because all data after that last snapshot would have been lost due to the systems breakdown. Getting back all that data would take a long time – time that could be better spent on more productive and profitable endeavors.
But what if you didn’t have to lose all the completed work? What if there was a way to improve RPOs by 10,000x? I knew that would get your attention.
The entire range of and old and new data protection solutions has attempted to tackle this challenge. Upgrades to traditional backup applications, newer VM-focused products as well as copy data management solutions all offer abilities to capture data changes in one way or another. However, practical RPOs are still mostly hours apart, with considerable expense on additional hardware, software and administration overhead.
Is there a solution better than traditional Storage Snapshots?
In 2015, Reduxio introduced BackDating™. It is essentially a built-in data protection technology that takes RPO’s down to 1 (yes, one) second. BackDating™ contrasts with snapshots in various ways–it’s continuously protecting the data, its use doesn’t call for a schedule, and it doesn’t require consistency groups since all data sets are consistent to a particular second.
BackDating™ today helps in seamless and instantaneous recovery from human errors, computer software failures, and malware–issues that for sure, all of us are constantly facing in our respective organizations.